CN102566555A - Major appliance work state monitoring method based on pattern recognition - Google Patents
Major appliance work state monitoring method based on pattern recognition Download PDFInfo
- Publication number
- CN102566555A CN102566555A CN2012100330144A CN201210033014A CN102566555A CN 102566555 A CN102566555 A CN 102566555A CN 2012100330144 A CN2012100330144 A CN 2012100330144A CN 201210033014 A CN201210033014 A CN 201210033014A CN 102566555 A CN102566555 A CN 102566555A
- Authority
- CN
- China
- Prior art keywords
- sample
- work state
- characteristic
- state monitoring
- monitoring method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention discloses a major appliance work state monitoring method based on pattern recognition. The major appliance work state monitoring method comprises the steps of (1) continuously acquiring multi-parameter signals when major appliances are in a normal work state in an online manner; (2) conducting manual classification, pretreatment and characteristic extraction on the work state in an offline manner, establishing a primitive characteristic sample database, wherein the characteristics are multi-dimensional characteristics which take current harmonics as the main and take amplitudes, phase positions and the like as the auxiliary; (3) utilizing the Euclidean distance algorithm to obtain an initial standard sample in an offline manner; (4) matching a testing sample and the initial standard sample in an offline manner, adopting a recycling search method to optimize the characteristics, and establishing a standard sample database; (5) conducting online test and automatic staging, and extracting the signal characteristics to obtain a sample to be tested; and (6) matching and recognizing the sample to be tested and the standard sample database in an online manner, so that the automatic appliance state classification can be realized. The method provided by the invention is suitable for online running work state monitoring, delivery quality control, failure prediction and diagnosis, appliance performance optimization control and the like of the major appliances.
Description
[technical field]
The present invention relates to the white domestic appliances field, be specifically related to a kind of duty of utilizing the method processing of pattern-recognition and analyzing white domestic appliances.Be applicable to health monitoring, the quality inspection of household electrical appliances combination property, network home appliance of household electrical appliances etc.
[background technology]
White domestic appliances have obtained people's more concern as the alternative people's housework and the electrical equipment that improves living standard of bettering people's living environment, and the white domestic appliances intellectuality will become the development trend of white domestic appliances.
At present, the white domestic appliances networked function is mainly reflected on Long-distance Control or the communication function, its remote fault diagnosis and service function relatively a little less than; In addition, household electrical appliances dispatch from the factory and often simple function are detected item by item when detecting, and the complete machine comprehensive function are detected lack effective means.It is to address the above problem the gordian technique that must capture that the duty of household electrical appliances is monitored in real time.
[summary of the invention]
The technical matters that the present invention will solve is the deficiency that overcomes prior art, and a kind of white domestic appliances work state monitoring method based on pattern-recognition is provided.
The technical scheme that the present invention adopts is that a kind of white domestic appliances work state monitoring method based on pattern-recognition comprises that step is following:
Step 1: the multi-parameter signals under the presence during continuous acquisition white domestic appliances normal operating conditions;
Step 2: under off-line state, duty is carried out the manual sort, pre-service, primitive character sample storehouse is set up in feature extraction;
Step 3: under off-line state, utilize the Euclidean distance algorithm to obtain the primary standard sample; Said primary standard sample be meant any sample in the training sample database to the sample minimum of other sample of equal state apart from sum;
Step 4: under off-line state,, optimize characteristic according to discrimination and generate master sample, form standard sample database by some master samples with test sample book and primary standard sample matches;
Step 5: on-line testing reaches automatically by stages, extracts signal characteristic, obtains sample to be tested;
Step 6: under the presence master sample in sample to be tested and the standard sample database is mated respectively, discerns, realize that tame electricity condition classifies automatically.
As preferably, be to adopt technology of frequency tracking in the above-mentioned steps 1 to being that main white domestic appliances working status parameter is gathered with the parameters of electric power, promptly follow the tracks of mains frequency and carry out synchronous acquisition.
As preferably, manual sort described in the step 2 is based under the program and mode of operation of known set, through observation of use senses and comparison of wave shape, is mapped different characteristic waveform and state; Said pre-service mainly is meant carries out denoising, zero passage detection etc. to signal; Said feature extraction is that to adopt the method for time and frequency zone analysis that digital signal is carried out extracting with current harmonics after the pre-service be main; Amplitude, phase place, power, peak factor, waveform trend, noise spectrum etc. are the multidimensional eigenmatrix of assisting; This multidimensional eigenmatrix is the primitive character sample, and the certain characteristics sample has been formed primitive character sample storehouse.
As preferably, the foundation of standard sample database described in the step 4 comprises following concrete steps:
(1) separating primitive character sample storehouse is training sample database and test sample book storehouse; Said training sample database comprises all training samples, and the test sample book storehouse comprises all test sample books, and be divided into several etc. the sample group of quantity;
(2) train from training sample through the Euclidean distance method and obtain the primary standard sample, form initial sample storehouse by some primary standard samples;
(3) all test sample book groups and primary standard sample are carried out based on relevant matching test, calculate the discrimination of each test sample book group and the average recognition rate of all groups;
(4) take the cyclic search method to optimize characteristic, search comprises all characteristics combination;
(5) select average recognition rate the highest and more than or equal to the characteristics combination of particular value S as optimizing characteristic, said particular value S ∈ [0.95,1); If the highest average recognition rate is less than S, then change or increase new feature, repeating step (4), until average recognition rate reach or greater than S till, the dimension of the sample that settles the standard and characteristic parameter form and optimize characteristic;
(6) set up the corresponding relation that relative optimization criteria sample and ID identification code are the duty coding, generate standard sample database.
As preferably, obtain sample to be tested described in the step 5 and comprise following concrete steps:
(1) online measured signal of obtaining is carried out based on the feature extraction of optimizing characteristic;
(2) utilize the similarity principle to carry out by stages on-line automatic to adjacent two waveforms.
As preferably, above-mentioned steps 6 is to utilize the method for simple crosscorrelation that sample to be tested and standard state sample storehouse are mated, and finds out corresponding ID identification code, confirms the duty classification.
The invention has the beneficial effects as follows:
1, monitors the household electrical appliances duty in real time, for network home appliance failure prediction and fault diagnosis provide basic data, for the household electrical appliances combination property detects and estimate the service that provides.
2, the method that proposes based on the present invention, but the real-time analysis control model to the influence of the key parameter of energy consumption, noise, thereby optimizer final optimization pass man electrical property (like energy-saving and noise-reducing).
3, have the most of power-type low-voltage electrical apparatus especially versatility of white domestic appliances.
[description of drawings]
Below in conjunction with accompanying drawing and embodiment the present invention is done further detailed explanation.
Fig. 1 is based on the white domestic appliances work state monitoring method process flow diagram of pattern-recognition.
Fig. 2 household electrical appliances work sequence figure.
Fig. 3 optimizes the sample dimension and sets up the standard sample database process flow diagram.
Fig. 4 gathers the current waveform figure of dissimilar household electrical appliances different conditions;
Wherein, 4-1 is the current waveform figure of certain model air-conditioning work state 1;
4-2 is the current waveform figure of certain model air-conditioning work state 2;
4-3 is the current waveform figure of certain model work state of washing machine 1;
4-4 is the current waveform figure of certain model work state of washing machine 2;
4-5 is the current waveform figure of certain model working station of microwave oven 1;
4-6 is the current waveform figure of certain model working station of microwave oven 2;
4-7 is the current waveform figure of certain model working station of microwave oven 3;
4-8 is the current waveform figure of certain model working station of microwave oven 4;
4-9 is the current waveform figure of certain model working station of microwave oven 5;
4-10 is the current waveform figure of certain model working station of microwave oven 6.
Current-voltage waveform comparison diagram under the micro-wave oven different operating state that Fig. 5 collected.
The recognition result comparison diagram of Fig. 6 model micro-wave oven state 6.
[embodiment]
As shown in Figure 1, comprise the following steps: based on the white domestic appliances work state monitoring method of pattern-recognition
Step 1: utilize technology of frequency tracking to gather tested white domestic appliances working status parameter information.China's electrical network standard frequency is 50HZ ± 0.2Hz.Be the synchronized sampling of bonding cycle N point (128 point), adopt N frequency multiplication technology of frequency tracking to gather voltage, current parameters.Temperature variation is slow, adopts low frequency to sample.Noise frequency is higher, adopts the high frequency Sampling techniques that it is gathered.
Step 2: the duty classification manual sort under off-line state accomplish.The off-line manual sort; At first according to the state under the program setting pattern-hold time--sequence relation (as shown in Figure 2); By people's observation of use senses by the outer characteristic of the household electrical appliances duty under the programmed control, the variation of involving vibrations (like washing machine), noise (like washing machine, refrigerator), temperature (air-conditioning).Secondly, continuous parameters wave form varies such as the voltage of off-line household electrical appliances duty that ATE is collected, electric current, temperature, noise are carried out waveform pattern signature analysis.Once more, carry out the time contrast through known program setting household electrical appliances duty variation and viewed waveform.Obtain some typical waveforms through a large amount of experiment statistics results and represent corresponding duty respectively, each state is chosen hundreds of groups of sample waveforms, sets up duty original sample storehouse, and each original sample wave period number equates.
Step 3: extract signal characteristic and set up primitive character sample storehouse.By among Fig. 4 shown in 4-1,4-2,4-3,4-4,4-5,4-6,4-7,4-8,4-9, the 4-10, the waveform that dissimilar household electrical appliances are corresponding is inequality, can be known by Fig. 5 again, and the waveform that household electrical appliances different operating state of the same type is corresponding is also inequality.So can distinguish the different operating state of dissimilar household electrical appliances according to waveform character.Consider that waveform has comprised abundant characteristic information on time domain and frequency domain, adopt time and frequency zone hybrid analysis method that waveform is analyzed.Through a large amount of experiments, the present invention extracts to comprise with current harmonics being main, and amplitude, phase place, power, peak factor, waveform trend, noise spectrum etc. are the characteristic parameter sample of assisting, and set up primitive character sample storehouse.Select for harmonic characteristic, get maximum 10 harmonic components and resultant distortion rate as characteristic.
Step 4: select the primary standard sample.Adopt the Euclidean distance method from training sample database, to obtain primary standard sample storehouse, each type state is selected a primary standard sample, this sample be in all similar samples with other samples sample minimum apart from sum.
Step 5: optimize the sample dimension and set up standard sample database.Wherein to set up flow process as shown in Figure 2 for standard sample database.
As shown in Figure 3, optimize the sample dimension and comprise following concrete steps:
(1) by calculating n original waveform sample in the step 2, after pre-service such as denoising, zero crossing detection etc., carries out feature extraction.Obtain n primitive character sample after the feature extraction, m training sample and (n-m) individual test sample book are arranged in the n feature samples.
(2) from training sample database, set up primary standard sample storehouse through the Euclidean distance method.
(3) test sample book and primary standard sample are carried out based on relevant matching test, calculate discrimination;
(4) take the cyclic search method to optimize characteristic.The finite dimension characteristic is adopted coupling respectively such as full dimension, combined type dimensionality reduction, single characteristic.
(5) select average recognition rate the highest and more than or equal to 0.99 characteristics combination as optimizing characteristic; If the highest discrimination is less than 0.99; Then change or the increase new feature, repeating step (4) is till discrimination is more than or equal to 0.99; The settle the standard dimension and the characteristic parameter of sample form and optimize characteristic;
(6) set up the corresponding relation that relative optimization criteria sample and ID identification code are the duty coding, generate standard sample database;
Step 6: work wave is by stages on-line automatic.Under the household electrical appliances running status, gather and calculate in real time the correlativity of adjacent two waveforms, confirm the state separation.Concrete steps are following:
(1) be a cycle with N point, first cycle of continuous acquisition leaves among the interim array a [N], and second cycle leaves among the interim array b [N].
(2) a [N] carries out related operation with b [N], according to the similarity degree of the represented waveform of resulting these adjacent two arrays of related coefficient judgement, then to a [N] zero clearing.
(3) the 3rd cycles are that N point after the starting point is stored in the array b [N] that deposits with last cycle acquired signal among the array a [N] and calculates correlativity with the 2N+1 point, to array b [N] zero clearing.
(4) repeat abovementioned steps (1), step (2) and step (3).Begin again to appear and jump out circulation when increasing trend suddenly when the coefficient of autocorrelation of adjacent two waveforms presents the trend of die-offing and reaches relatively minimum, and with the corresponding state separation of constantly confirming as of current cycle.
5, be the relative time benchmark with said separation cycle; Select its before and after symmetrical the adjacent the 1st or the individual complete cycle wave-wave shape of k (k >=1) as transition state (statistical law that changes speed according to the different conditions of different household electrical appliances is confirmed the k value), the continuous wave between adjacent two transition states is a steady state (SS).
6, initial, the end sampling point sequence number of real time record transition state or steady state (SS) multiply by the SI, obtain the working time of transition or steady state (SS).Arrange according to the order of sequence, form continuum of states-sequential chart (this fashion does not obtain state ID).So far, realize waveform automatically by stages.
Step 7: obtain the steady state (SS) waveform according to step 6, select time is that 4 continuous waves of intermediate point carry out pre-service, optimize feature extraction, deposits array in as sample to be tested.
Step 8: pattern match.Utilize the method for simple crosscorrelation to calculate sample to be tested and standard sample database matching degree.Be about to sample to be tested and all samples of standard sample database relevant matches one by one, seeking degree of correlation the maximum is matching status.
Step 9: draw the state ID identification code that matches in the standard sample database according to matching degree, confirm tested state.Be illustrated in figure 6 as the recognition result of certain model working station of microwave oven 6, by finding out in the histogram, the histogram of ID6 correspondence is near 1.
Claims (6)
1. the white domestic appliances work state monitoring method based on pattern-recognition is characterized in that, comprises that step is following:
Step 1: the multi-parameter signals under the presence during continuous acquisition white domestic appliances normal operating conditions;
Step 2: under off-line state, duty is carried out manual sort, pre-service, feature extraction, set up primitive character sample storehouse;
Step 3: under off-line state, utilize the Euclidean distance algorithm to obtain the primary standard sample; Said primary standard sample be meant any sample in the training sample database to the sample minimum of other sample of equal state apart from sum;
Step 4: under off-line state,, optimize characteristic according to discrimination and generate master sample, form standard sample database by some master samples with test sample book and primary standard sample matches;
Step 5: on-line testing reaches automatically by stages, extracts signal characteristic, obtains sample to be tested;
Step 6: under the presence master sample in sample to be tested and the standard sample database is mated respectively, discerns, realize that tame electricity condition classifies automatically.
2. white domestic appliances work state monitoring method according to claim 1; It is characterized in that; Be to adopt technology of frequency tracking in the said step 1, promptly follow the tracks of mains frequency and carry out synchronous acquisition being that main white domestic appliances working status parameter is gathered with the parameters of electric power.
3. white domestic appliances work state monitoring method according to claim 1; It is characterized in that; Manual sort described in the step 2 is based under the program and mode of operation of known set, through observation of use senses and comparison of wave shape, is mapped different characteristic waveform and state; Said pre-service mainly is meant carries out denoising, zero passage detection etc. to signal; Said feature extraction is that to adopt the method for time and frequency zone analysis that digital signal is carried out extracting with current harmonics after the pre-service be main; Amplitude, phase place, power, peak factor, waveform trend, noise spectrum etc. are the multidimensional eigenmatrix of assisting; This multidimensional eigenmatrix is the primitive character sample, and the certain characteristics sample has been formed primitive character sample storehouse.
4. white domestic appliances work state monitoring method according to claim 1 is characterized in that, the foundation of standard sample database described in the step 4 comprises following concrete steps:
(1) separating primitive character sample storehouse is training sample database and test sample book storehouse; Said training sample database comprises all training samples, and the test sample book storehouse comprises all test sample books, and be divided into several etc. the sample group of quantity;
(2) train from training sample through the Euclidean distance method and obtain the primary standard sample, form initial sample storehouse by some primary standard samples;
(3) all test sample book groups and primary standard sample are carried out based on relevant matching test, calculate the discrimination of each test sample book group and the average recognition rate of all groups;
(4) take the cyclic search method to optimize characteristic, search comprises all characteristics combination;
(5) select average recognition rate the highest and more than or equal to the characteristics combination of particular value S as optimizing characteristic, said particular value S ∈ [0.95,1); If the highest average recognition rate is less than S, then change or increase new feature, repeating said steps (4), until average recognition rate reach or greater than S till, the dimension of the sample that settles the standard and characteristic parameter form and optimize characteristic;
(6) set up the corresponding relation that relative optimization criteria sample and ID identification code are the duty coding, generate standard sample database.
5. white domestic appliances work state monitoring method according to claim 1 is characterized in that, obtains sample to be tested described in the step 5 and comprises following concrete steps:
(1) online measured signal of obtaining is carried out based on the feature extraction of optimizing characteristic;
(2) utilize the similarity principle to carry out by stages on-line automatic to adjacent two waveforms.
6. white domestic appliances work state monitoring method according to claim 1 is characterized in that, said step 6 is to utilize the method for simple crosscorrelation that sample to be tested and standard state sample storehouse are mated, and finds out corresponding ID identification code, confirms the duty classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100330144A CN102566555B (en) | 2012-02-10 | 2012-02-10 | Major appliance work state monitoring method based on pattern recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100330144A CN102566555B (en) | 2012-02-10 | 2012-02-10 | Major appliance work state monitoring method based on pattern recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102566555A true CN102566555A (en) | 2012-07-11 |
| CN102566555B CN102566555B (en) | 2013-12-04 |
Family
ID=46412199
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012100330144A Active CN102566555B (en) | 2012-02-10 | 2012-02-10 | Major appliance work state monitoring method based on pattern recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102566555B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799108A (en) * | 2012-07-27 | 2012-11-28 | 中国海洋大学 | Energy consumption depolymerization method based on low frequency energy consumption information and home energy management system |
| CN104135407A (en) * | 2014-08-12 | 2014-11-05 | 中国联合网络通信集团有限公司 | Electrical apparatus operational monitoring method, server and system |
| CN104198842A (en) * | 2014-08-11 | 2014-12-10 | 无锡和晶信息技术有限公司 | Whole machine state detection assisting method based on harmonic current and power |
| CN104597335A (en) * | 2013-10-30 | 2015-05-06 | 海尔集团公司 | Working condition monitoring system and monitoring method for household appliances |
| CN105467975A (en) * | 2015-12-29 | 2016-04-06 | 山东鲁能软件技术有限公司 | Equipment fault diagnosis method |
| CN107179455A (en) * | 2017-04-27 | 2017-09-19 | 华中科技大学 | A kind of real-time electrical appliance recognition and system based on edge features vector model |
| CN109584232A (en) * | 2018-11-28 | 2019-04-05 | 成都天衡智造科技有限公司 | Equipment use state on-line monitoring method, system and terminal based on image recognition |
| CN110207967A (en) * | 2019-06-13 | 2019-09-06 | 大连海事大学 | A kind of state identification method and system based on wavelet pack energy feature and cross-correlation |
| CN111368928A (en) * | 2020-03-06 | 2020-07-03 | 普迪飞半导体技术(上海)有限公司 | Wafer pattern matching method and device, electronic equipment and storage medium |
| CN112631242A (en) * | 2020-12-07 | 2021-04-09 | 国网四川省电力公司电力科学研究院 | Power abuse analysis method and device for household electrical equipment |
| CN116755414A (en) * | 2023-08-22 | 2023-09-15 | 山东新巨龙能源有限责任公司 | Ore mining equipment supervision system based on Internet of things |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5123057A (en) * | 1989-07-28 | 1992-06-16 | Massachusetts Institute Of Technology | Model based pattern recognition |
| US20080144072A1 (en) * | 2006-06-23 | 2008-06-19 | Iregistry Pty Ltd. | Validation system and method therefor |
| US8055062B2 (en) * | 2007-11-13 | 2011-11-08 | Sony Corporation | Information processing apparatus, information processing method, and program |
-
2012
- 2012-02-10 CN CN2012100330144A patent/CN102566555B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5123057A (en) * | 1989-07-28 | 1992-06-16 | Massachusetts Institute Of Technology | Model based pattern recognition |
| US20080144072A1 (en) * | 2006-06-23 | 2008-06-19 | Iregistry Pty Ltd. | Validation system and method therefor |
| US8055062B2 (en) * | 2007-11-13 | 2011-11-08 | Sony Corporation | Information processing apparatus, information processing method, and program |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799108B (en) * | 2012-07-27 | 2014-12-17 | 中国海洋大学 | Energy consumption depolymerization method based on low frequency energy consumption information and home energy management system |
| CN102799108A (en) * | 2012-07-27 | 2012-11-28 | 中国海洋大学 | Energy consumption depolymerization method based on low frequency energy consumption information and home energy management system |
| CN104597335A (en) * | 2013-10-30 | 2015-05-06 | 海尔集团公司 | Working condition monitoring system and monitoring method for household appliances |
| CN104198842A (en) * | 2014-08-11 | 2014-12-10 | 无锡和晶信息技术有限公司 | Whole machine state detection assisting method based on harmonic current and power |
| CN104135407A (en) * | 2014-08-12 | 2014-11-05 | 中国联合网络通信集团有限公司 | Electrical apparatus operational monitoring method, server and system |
| CN105467975A (en) * | 2015-12-29 | 2016-04-06 | 山东鲁能软件技术有限公司 | Equipment fault diagnosis method |
| CN105467975B (en) * | 2015-12-29 | 2018-08-03 | 山东鲁能软件技术有限公司 | A kind of equipment fault diagnosis method |
| CN107179455B (en) * | 2017-04-27 | 2019-04-23 | 华中科技大学 | A kind of real-time electrical appliance recognition and system based on edge features vector model |
| CN107179455A (en) * | 2017-04-27 | 2017-09-19 | 华中科技大学 | A kind of real-time electrical appliance recognition and system based on edge features vector model |
| CN109584232A (en) * | 2018-11-28 | 2019-04-05 | 成都天衡智造科技有限公司 | Equipment use state on-line monitoring method, system and terminal based on image recognition |
| CN110207967A (en) * | 2019-06-13 | 2019-09-06 | 大连海事大学 | A kind of state identification method and system based on wavelet pack energy feature and cross-correlation |
| CN110207967B (en) * | 2019-06-13 | 2020-12-01 | 大连海事大学 | State identification method and system based on wavelet packet energy characteristics and cross correlation |
| CN111368928A (en) * | 2020-03-06 | 2020-07-03 | 普迪飞半导体技术(上海)有限公司 | Wafer pattern matching method and device, electronic equipment and storage medium |
| CN111368928B (en) * | 2020-03-06 | 2023-04-07 | 普迪飞半导体技术(上海)有限公司 | Wafer pattern matching method and device, electronic equipment and storage medium |
| CN112631242A (en) * | 2020-12-07 | 2021-04-09 | 国网四川省电力公司电力科学研究院 | Power abuse analysis method and device for household electrical equipment |
| CN116755414A (en) * | 2023-08-22 | 2023-09-15 | 山东新巨龙能源有限责任公司 | Ore mining equipment supervision system based on Internet of things |
| CN116755414B (en) * | 2023-08-22 | 2023-11-07 | 山东新巨龙能源有限责任公司 | Ore mining equipment supervision system based on Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102566555B (en) | 2013-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102566555B (en) | Major appliance work state monitoring method based on pattern recognition | |
| CN107025365B (en) | A kind of non-intruding load discrimination method for user side | |
| CN110956220B (en) | Non-invasive household appliance load identification method | |
| CN103645425B (en) | High-voltage cable insulation defect partial discharge on-line monitoring diagnosis method | |
| CN103197218B (en) | A kind of high-voltage cable insulation defect partial discharge electrification detection diagnostic method | |
| CN110514889A (en) | A kind of method and system of non-intrusion type household electricity remained capacity | |
| CN105067966B (en) | The low-voltage alternating-current fault arc detection method of feature based modal components energy spectrometer | |
| CN112505511B (en) | Non-invasive low-voltage fault arc detection and positioning method and system | |
| CN106441547A (en) | Transformer vibration monitoring method and apparatus | |
| CN111382789B (en) | Power load identification method and system based on machine learning | |
| CN106646026A (en) | Non-intrusive household appliance load identification method | |
| CN104198898A (en) | Local discharge development process diagnosis method based on pulse-train analysis | |
| CN111242391A (en) | Machine learning model training method and system for power load identification | |
| CN105372541A (en) | Household appliance intelligent set total detection system based on pattern recognition and working method thereof | |
| CN109782086B (en) | Non-intrusive load identification method based on multi-dimensional signal analysis | |
| CN105676028B (en) | A kind of resident load electricity consumption recognition methods based on template matches filtering | |
| CN111308260B (en) | Electric energy quality monitoring and electric appliance fault analysis system based on wavelet neural network and working method thereof | |
| CN109449919A (en) | A kind of non-intrusion type analysis of power consumption load method and device | |
| CN103675544A (en) | Power system fault signal detection and waveform identification method based on optimization algorithm | |
| CN111327118B (en) | Non-invasive power load identification method | |
| Hernandez et al. | Development of a non-intrusive load monitoring (nilm) with unknown loads using support vector machine | |
| CN102799108B (en) | Energy consumption depolymerization method based on low frequency energy consumption information and home energy management system | |
| WO2022100046A1 (en) | Mechanical transmission fault detection method based on zero-value ratio spectrum feature | |
| Bucci et al. | A non-intrusive load identification system based on frequency response analysis | |
| CN105891635A (en) | Digital signal processing-based electric appliance type identification device and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee |
Owner name: ANHUI JIANZHU UNIVERSITY Free format text: FORMER NAME: ANHUI BUILDING INDUSTRY COLLEGE |
|
| CB03 | Change of inventor or designer information |
Inventor after: Gao Cuiyun Inventor before: Gao Cuiyun Inventor before: Liu Ming Inventor before: Han Ru Inventor before: Li Wenjing |
|
| CB03 | Change of inventor or designer information | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: LI BAODONG ZHU HAIFEI ZOU JIANPING CHEN WEI TO: LI BAODONG |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 230601 Hefei economic and Technological Development Zone, Anhui, No. 292 Ziyun Road Patentee after: Anhui University of Architecture Address before: 230601 Hefei economic and Technological Development Zone, Anhui, No. 292 Ziyun Road Patentee before: anhui university of architecture |
|
| CP01 | Change in the name or title of a patent holder |